Thoughts on turbocharger behaviour

Competition engines and 'live' projects only. Good photos to illustrate your post are expected.
Julian
Posts: 181
Joined: June 22nd, 2006, 6:45 pm
Location: Manchester, UK
Contact:

Turbine in isolation

Post by Julian »

First all apologies on what is a somewhat rambling post - I hope everyone can make sense of it or if not someone more elequent that myself can translate it :wink:

Considering the turbine in isolation provides a useful insight but without considering the engine as a whole it doesn't resolve the sort of matching issues (at any level) that we are likely to face.

Just changing the turbine on any engine is destined to a poor choice of action simply because everything else is tailored to the performance of the original not least of all the processing of thermal waste.

In terms of extracting a broader range of performance it strikes me that the turbocharger in general is misused. The optimal performance envelope for any given turbo is always limited and while we can attempt to broaden that envelope it is very hard to achieve and potentially very expensive - for the grassroots tuner likely to be beyond the realms of viable endeavour.

Two strategies spring to mind that can assist in mastering the potential of the turbo and developing a more useful output. The first of these is already in use with the VAG turbo petrol engines and relies on the use of an electronically controlled valve that effectively limits the dynamic compression ratio. The bane of turbo petrol engines is the need for a low CR (as already stated above) but this means that without the boost the engine is effectively just a low compression engine producing little torque. The most boost the less static compression the engine needs to prevent component failure. In a race engine that spends all of its time under load this is fine but suffers horrendously off boost (the top gear demonstration of how bad the Evo IX FQ behaved was a very graphic illustration). For a more versatile engine the static CR needs to be kept as high as possible though. A smaller turbo (in terms of mass flow) would lead to the engine staying within limits under dynamic compression but running out of puff at higher revs (which leads us back to why diesel turbos are so good). A larger turbo capable of providing sufficient mass flow across the desired rev range would be ideal but could potentially produce an excess while operating in its optimum performance envelope. The control valve wastes this excess keeping the boost under control and the dynamic compression where the engine can operate without risk of destroying itself.

The second possiblity is a more recent development for the road car and is apparent with the latest generation of engines due in the next year and gives a new meaning to dynamic compression ratio. The engine is capable of internally wasting excess compression and effectively performs the same task as the valve used by VAG. With the hardware integrated into the engine itself the need for an external valve is lost. This variable static compression ratio mechansim has the potential to assist with off-boost running to maintain more linear performance. Combined with the now ubiquitous variable cam timing it is possible to envisage an engine that can provide optimal performance in all respects - economic, low compression and lean burn for cruising with high performance on tap when required.

In terms of changing the requirements of matching the turbo to the engine little has changed (the engineer still has to make sure that everything works in unison) but the expectations of the consumer is more likely to be fulfilled. For fast road and race track applications both solutions have great potential (the variable compression ratio is not limited to forced induction making it of even greater interest). The reality of converting an existing design to variable compression is simply that it is unviable (for now) but the ability to effectively map boost is much more realistic.
sumplug
Posts: 234
Joined: June 25th, 2006, 10:25 am
Location: Banned 4th Oct 07 by GC
Contact:

Post by sumplug »

julian, you have missed the company that brought turbocharging to the masses. that is SAAB. there 900/9000 models had in the early 1980's , a control box called the Automatic performance control or APC for short. this allowed a compression of 9-1 and would control the turbo boost accordingly using a knock sensor. the APC was gradually updated and was a wonderful piece of kit that worked beautifully right into the mid 1990's. 8)
Kev Rooney
Posts: 45
Joined: June 26th, 2006, 7:29 am
Location: Portsmouth

Post by Kev Rooney »

Look to VW's new twincharger motors for the answer. Supercharger that provides instant bottom end grunt and it's then bypassed as the turbo is providing the top end. They run something like 40 psi and this is a a normal road going car.

I'm still researching ( and assembling ) info for my turbo conversion on a 1600 Fiat .Currently making a manifold to take a more suitable turbo that will provide power from 1500 rpm upwards. Not an unltimate power machine but a torque monster which is what's required for a all round street engine.
sumplug
Posts: 234
Joined: June 25th, 2006, 10:25 am
Location: Banned 4th Oct 07 by GC
Contact:

Post by sumplug »

for the 1600, run a G20 turbo with APC from a saab to control it.
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

Kev hi

don't turbo it - supercharge it, much nicer torque curve. Like this one I am doing for Kev Rendle, 131 TC 185 motor, Vx blower, 3A GC inlet cam, fully flowed, Vx head, race pistons, dry sump, steel fw, for his stunning and much modded Tiumph Spitfire..

BTW Kev Rendle Spitfire - photos pleeesa!

Not enough non-F/L features on this site, no ho!
Attachments
KR eng sump & cambelt on front view.jpg
KR eng sump & cambelt on front view.jpg (83.49 KiB) Viewed 13982 times
Acki

Post by Acki »

What's a nicer troque curve? Like a pacer? :D
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

ha ha - now you're testing my knowledge of English! again in German!

GC
Acki

Post by Acki »

Pacer = engine without turbo etc.

Again in german?

Was ist eine nette Drehmomentkurve?
Wie ein Saug-Motor :D
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

yes,

similar to a normally aspirated torque in shape, without the lag phase of the turbo, flatter than any other torque curve, and more of it everywhere. A well tuned supercharged engine will every turbo for dead, boost for boost anyhow. The problem is that most superchargers struggle to produce much over 15 psi boost, over that turbos win out, hence the Delta S4...
Attachments
Simon Bray No2 car Vx 2 liter had great fun at Goodwood in among Integrales
Simon Bray No2 car Vx 2 liter had great fun at Goodwood in among Integrales
copyapril06goodwood002.jpg (74.99 KiB) Viewed 13956 times
Simon Bray's Vx unit.
Simon Bray's Vx unit.
blower setup top_01.JPG (33.42 KiB) Viewed 13954 times
Last edited by Guy Croft on July 7th, 2006, 10:31 am, edited 3 times in total.
Acki

Post by Acki »

15 psi ~ 1 bar? Hmm I will drive 1.7 bar with kkk charger :)
Maybe a supercharger has more power at low revs but after the turbo lag (3.000rpm) I have more power than a supercharger.
I drive a Fiat, a italian car, when you drive it with a turbo charger a turbo lag is a must ;) It's the italian mentality :P
I like the kick when the turbo begins to boost. :)

I think you must drive both to make a choice what you wants to drive/have :)
Testament
Posts: 101
Joined: June 22nd, 2006, 7:47 pm
Location: Taupo, New Zealand
Contact:

Post by Testament »

These vent valve systems to limit boost and dynamic compression ratio doesn't sit that well from my engineering perspective. One of the main benefits of a turbocharger is improved efficiency, by venting boost once you have done work to produce it is more wasteful than not doing the work in the first place and letting the exhaust energy go out the tailpipe. The variable nozzle turbine (VNT) units used on new diesel engines and that latest porsche turbo are a better method (albeit more complicated) of widening a turbochargers range of effectiveness.

With the improvements in electronics and control systems in the last 10 years or so the compression ratios of factory turbocharged cars has been rising, and low down performance improving markedly. I don't know which Evo you are talking about exactly, I know there are a number of FQ 'modified' versions you get in the UK, but the factory Japanese spec ones we have in NZ do not have any huge hole down low since at least the Evo 7.
Julian
Posts: 181
Joined: June 22nd, 2006, 6:45 pm
Location: Manchester, UK
Contact:

Post by Julian »

You'd have to explain that particular one to me - the idea is not so much to waste work done since the boost is still there, only the excess is disposed of and since the boost level remains constant only a small part of the work is lost, in essence you are removing the load on the boost compression above that critical (controlled) level.

With regards the Evo (it was the 8 I believe) that came in a number of flavours. The most obvious of these is the MR FQ430 (430 for 430bhp at the flywheel - an impressive figure for a 2 litre road car). The problem was demonstrated very nicely on Top Gear a couple of years back. If you keep the car on boost it will accelerate like a demon but as soon as the boost is lost (by backing off or trying to accelerate hard "off-cam") the lag was hideous - the sort of thing that really early turbo cars used to suffer from in the days of the Starion. The problem is that in order to obtain the 430bhp they had resorted to using a turbo that only worked at the top end of the engine's performance envelope. [ I could dig out the relevant information on the Garrett website but it will have to wait until I can come back and edit this post ]. Outside of the desired performance the turbo is simply a huge drain on the energy developed. The static compression ratio had to be set very low to cope with the high boost used so it all gets in a very sticky mess.
Testament
Posts: 101
Joined: June 22nd, 2006, 7:47 pm
Location: Taupo, New Zealand
Contact:

Post by Testament »

Ok so that ws the FQ300 or similar. The factory cars have 9:1 compression and the japanese limit '280hp' but are probably 300+hp in reality at the 18psi factory boost setting.

The boost level may be the same, but as you are saying excess air is still being disposed of. Thermodynamically its more efficient to put less energy into the turbine and not dispose of any of the compressed inlet air.
Julian
Posts: 181
Joined: June 22nd, 2006, 6:45 pm
Location: Manchester, UK
Contact:

Post by Julian »

Much as we'd like to believe otherwise we all know you don't get something for nothing. If you gain in one area you lose somewhere else.

In terms of achieving optimum efficiency I am not worrying too much as this is for outright performance - drag racing and circuit racing. The important feature is that the engine generates boost quickly (ie without lag) and across a broad range of revs. In order to do this the compression ratio has to stay quite high by comparison. To prevent an internal failure you have to restrict the boost to a calculated limit.

One of the key features of a turbocharger is the exponential increase in power while on boost as revs increase. This is great if you want massive peak power but as already realised decades ago it leads to all sorts of problems on the road. It isn't very often you hear about turbo cars ramming the car infront when the boost "runs-away", you no longer have to time your overtake maneuvers to compensate for the sudden delivery or the lag before it arrives.

For racing purposes the natural unchecked behaviour is quite desirable provided you can drive around the inherent problems while cornering. What you really want though is something with the versatility of supercharging but operating at the top end of the rev range and as simple as you can make it. The solution provided for my engine covers all of these points to provide (according to dyno tests) about 300bhp from the 1400 Fiat turbo engine at 1.6 bar or 250bhp at 1.2 bar. The delivery isn't peaky - the mule for this engine has been run (on drag strips) against the latest Impreza WRX STi and won by a noticible margin, it then went on to do the same trick against a Skyline GTR R34 - both of these cars driven by skilled drivers, both cars have significant advantages in terms of torque, power and traction although they do lose out on weight.

Audi took great interest in the car and it was given red carpet treatment at their development centre near the nurburgring. The owner (perhaps foolishly) handed the keys over to their test driver and he described it as one of the fastest cars he's ever driven (and one of the worst handling).

The downside is that after all the work is done and the car racks up a fuelling bill the cost is far from small. It also takes a fair degree of skill to map the engine in the first place. The original test engine though ran non-stop for a week under constant loading and only gave up when the operator got bored and allowed it to rev to silly levels - whereupon the oil pump drive shaft snapped and the rest of the engine quickly followed pretty much destroying the test cell walls. The mule engine blew the turbo up after repeated power runs on a rolling road while trying to record a figure over 320bhp at full boost (highest permitted by the control valve was 2bar) - the engine itself was fine but the turbocharger was looking very sorry for itself.

None of these things are really desirable for a road car - the cost is too high for anything other than occasional usage and you certainly wouldn't want to keep paying for new turbochargers.

Anway - enough rambling for now. I'm running out of lunch break!
Julian
Posts: 181
Joined: June 22nd, 2006, 6:45 pm
Location: Manchester, UK
Contact:

Post by Julian »

I don't have all of the information to hand - the owner of the car was rather reluctant to spill all of the beans on how he destroyed the turbo on the day. It was on a rolling road so I wouldn't have expected the inlet temperature to be particularly favourable. A chargecooler packed with ice water would likely have improved the output quite nicely but I know the car wasn't equipped with anything like that - just a fairly moderately sized intercooler.

Just keep in mind that this is the old SOHC engine and while the head was modified to work in a turbo setup it was never engineered to operate this way. It doesn't benefit from a crossflow port arrangement or four valves per cylinder. Ultimately the head is just too restrictive - originally we had thought that we wouldn't get past 270bhp and 300bhp came as a pleasant surprise.

An equivalent 16v head might boost the potential by another 15% or maybe 20% but not much more.

The other thing to keep in mind is that the peak power isn't necessarily of that much interest - the engine configuration is to provide a broad range of torque. In terms of acceleration it is positively explosive in pretty much any gear and at any speed. Ultimately though the best of the performance is over and done with at about 6000rpm, the engine will rev happily to 9000rpm but you are much better off changing up a gear and taking advantage of the peak torque.

I wish I could give much more exact figures for the engine performance but ultimately I don't have any output graphs here and have to work from memory - I certainly don't want to start making the figures up. My race engine is still unfinished - as soon as it is running I will happily post up the results. If I can dig the older information from their respective owners i will post those too.
Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests